DEPARTMENT OF THE AIR FORCE
HEADQUARTERS AIR FORCE CIVIL ENGINEER SUPPORT AGENCY
4 NOV 2005
FROM:
HQ AFCESA/CES
139 Barnes Drive, Suite 1
Tyndall AFB, FL 32403-5319
SUBJECT: Engineering Technical Letter (ETL) 05-8: Use of Off-the-Shelf
Concrete Admixtures as Cold Weather Admixture Systems (CWAS).
1. Purpose. This ETL provides guidance to design, mix, place, and cure concrete in
below-freezing weather.
Note: The use of the name or mark of any specific manufacturer, commercial product,
commodity, or service in this ETL does not imply endorsement by the Air Force.
2. Application: All Air Force organizations with concrete construction, maintenance,
and repair responsibility.
2.1. Authority: Air Force policy directive (AFPD) 32-10, Air Force Installations and
Facilities, and Air Force instruction (AFI) 32-1023, Design and Construction
Standards and Execution of Facility Construction Projects.
2.2. Coordination: Major command (MAJCOM) pavement engineers.
2.3. Effective Date: Immediately.
2.4. Ultimate Recipients:
• Air Force MAJCOM pavement engineers.
• Base civil engineers (BCE), RED HORSE, and other units responsible for
design, construction, maintenance, and repair of concrete structures.
• U.S. Army Corps of Engineers (USACE) and Navy offices responsible for Air
Force design and construction.
3. Referenced Publications.
3.1. Air Force:
• AFPD 32-10, Air Force Installations and Facilities, available at
http://www.e-publishing.af.mil/
• AFI 32-1023, Design and Construction Standards and Execution of Facility
Construction Projects, available at http://www.e-publishing.af.mil/
3.2. Army:
• Engineer Research and Development Center, Cold Regions Research and
Engineering Laboratory Technical Report (ERDC/CRREL TR) 04-2,
APPROVED FOR PUBLIC RELEASE: DISTRIBUTION UNLIMITED
•
Extending the Season For Concrete Construction and Repair: Phase I—
Establishing the Technology, available at
http://www.crrel.usace.army.mil/techpub/CRREL_Reports/reports/TR04-2.pdf
ERDC/CRREL TR-05-9, Placing Antifreeze Concrete at Grand Forks Air
Force Base, available at
http://www.crrel.usace.army.mil/techpub/CRREL_Reports/reports/TR05-9.pdf
3.3. Industry:
• American Concrete Institute (ACI) 306-R88, Cold Weather Concreting,
available at http://www.concrete.org/BOOKSTORE/bkstr.asp
• American Society for Testing and Materials (ASTM) C33, Standard
Specification for Concrete Aggregates, available at http://www.astm.org
• ASTM C150, Standard Specification for Portland Cement, available at
http://www.astm.org
4. Acronyms and Terms:
ACI
- American Concrete Institute
AEA
- air-entraining admixture
ASTM
- American Society for Testing and Materials
BCE
- base civil engineer
C
- Celsius
CWAS
- cold weather admixture system
ERDC/CRREL - Engineer Research and Development Center, Cold Regions Research
and Engineering Laboratory
ETL
- Engineering Technical Letter
FHWA
- Federal Highway Administration
kg
- kilogram
3
kg/m
- kilograms per cubic meter
L
- liter
3
m
- cubic meter
MAJCOM
- major command
MB
- Master Builders, Inc.
mL
- milliliter
mm
- millimeter
NH
- New Hampshire
NMSA
- nominal maximum size aggregate
RED HORSE - Rapid Engineers Deployable - Heavy Operational Repair Squadron
Engineers
rpm
- revolutions per minute
TR
- Technical Report
USACE
- U.S. Army Corps of Engineers
w/c
- water/cement
WI
- Wisconsin
WRG
- W.R. Grace & Co.
2
5. Objective. This ETL provides information for combining ordinary concrete admixtures
into cold weather admixture systems (CWAS) that depress the freezing point of water to
at least –5°C and accelerate the hydration rate of Portland cement while at that low
temperature.
6. Background.
6.1. Currently, there are no commercially available admixtures that, when used
alone, will prevent fresh concrete from freezing at an internal temperature of –5°C.
Admixtures are available that allow concrete to gain strength at air temperatures
below 0°C, but these admixtures, when used at their recommended dosages, will not
prevent freezing. They promote strength gain by accelerating cement hydration,
which increases the rate of internally generated heat to help maintain concrete
temperatures above freezing until enough strength is developed to resist damage
from freezing.
6.2. This ETL presents the tools to design, mix, place, and cure concrete made with
combinations of commercial off-the-shelf admixtures in below-freezing weather.
Because standard practice places no limitations on the number of admixtures that
may be used in concrete—just on individual dosages—several admixtures were
combined to produce the desired low-temperature effect. This ETL is based on the
results of laboratory tests and field studies obtained between 1 October 2000 and 1
October 2003 under the Federal Highway Administration (FHWA) State Planning
and Research Pooled Fund Study TPF-5(003), Extending the Season for Concrete
Construction and Repair: Phase I—Establishing the Technology (ERDC/CRREL TR04-2), and on a demonstration of the resulting technology at Grand Forks Air Force
Base, North Dakota (ERDC/CRREL TR-05-9, Placing Antifreeze Concrete at Grand
Forks Air Force Base).
7. Definition of Cold Weather Admixture System (CWAS). CWAS is defined as any
commercial chemical admixture or combination of commercial chemical admixtures that
depresses the freezing point of mixing water and accelerates the hydration rate of
Portland cement in concrete. The term antifreeze admixture may be used
interchangeably with the term CWAS. Concrete containing a CWAS may be referred to
as antifreeze concrete.
8. Function of CWAS. The primary functions of CWAS are to prevent fresh concrete
from freezing at an internal temperature as low as –5°C and to promote strength gain at
that temperature at least as fast as normal concrete cured at 5°C. The secondary
function is to create an antifreeze concrete mixture that behaves like conventional
concrete at time of placement.
9. Materials.
9.1. Aggregate. Aggregates, both fine and coarse, that are approved for normal
concreting practice are acceptable for antifreeze concrete. The aggregates used to
3
develop this ETL were locally available, clean and well graded to produce workable
concrete mixtures, and were known to be durable. There are no size restrictions, but
the nominal maximum size coarse aggregate (NMSA) used to develop this ETL was
19 millimeters for all tests except the Rhinelander, Wisconsin, field trial
(ERDC/CRREL TR-04-2), which used a 38-millimeter NMSA coarse aggregate.
Typically, coarse aggregates used in pavements are larger than those used in other
concrete structures; the Rhinelander job was a pavement, while the other jobs were
not. The fine aggregates in all cases were natural sands meeting the specifications
of ASTM C33, Standard Specification for Concrete Aggregates.
9.2. Cement.
9.2.1. The testing for this ETL was limited to Portland cement. Though blended
cements such as fly ash, blast furnace slag, or silica fume combined with
Portland cement are common in general construction, they were not included
because it was decided that the antifreeze technology should be proven on
Portland cement first. Now that the antifreeze properties of admixtures for winter
concrete construction are proven, this technology should be expanded to include
other cements.
9.2.2. Portland cements that are specified for cold weather concreting are
acceptable for antifreeze concrete. Cements that conform to ASTM C150,
Standard Specification for Portland Cement, Type I and Type III, are usually
specified for winter concreting. This ETL allows Type I and Type II cements, plus
a combination Type I/II cement. Type III, high early strength cement, was not
studied because it is not widely available. The Type II cement, though it
generates less heat and at a slower rate than Type I, is permissible because the
high doses of accelerating admixtures used to formulate the CWAS should
overcome most tendencies for slow strength development. The Type II cement
has performed quite well in past studies and is recommended for use with the
admixtures approved in this ETL.
9.2.3. In today’s practice, obtaining high early age strength is advantageous in
cold weather to decrease the length of time that thermal protection is necessary.
This ETL focuses on that concept. The usual methods of achieving high early
strength are to use all Type III cement, use additional Type I cement, or use
chemical accelerators. One admixture formulation, WRG IV in the FHWA project
report (ERDC/CRREL TR-04-2), used a retarding admixture expressly to slow
down the slump loss characteristic of concretes made with high doses of
accelerating admixtures. Unfortunately, the retarder did little to solve the early
stiffening problem, but it delayed strength gain, which was viewed as a negative
consequence. However, for bridge deck applications, where delayed strength
gain may be desirable, a retarded mix such as WRG IV might prove useful. But,
because the focus of the studies leading up to this ETL was on rapid strength
gain, the WRG IV formulation was not included in this ETL.
4
9.2.4. To create antifreeze concrete with high early age strength, it is
recommended to use at least 363 kilograms per cubic meter (kg/m3) of Type I,
Type II, or Type I/II cement. The lowest cement factor used in this ETL is 392
kg/m3, while the highest is 476 kg/m3. Also, the water/cement (w/c) ratio should
be kept at or below 0.45 in accordance with current guidance for durable
concrete. Lower w/c ratios, besides further increasing durability and enhancing
strength gain, also lessen the amount of admixture needed for freezing point
depression.
9.3. Water. Water approved for making normal concrete can be used as mixing
water for making antifreeze concrete. Cold water, as explained later, should be used
to batch all antifreeze concrete mixtures to slow down slump loss. The
recommended initial temperature of the concrete is 10°C ± 5°C.
9.4. Chemical Admixtures. To avoid compatibility problems among admixtures and
limit the possible combinations of admixtures to investigate, this ETL concentrates
on the product lines of W.R. Grace & Co. (WRG) and Master Builders, Inc. (MB).
(Admixtures from other sources could similarly be used, but they should be tested in
laboratory and field trials to be sure they produce the desired results.) Table 1 shows
six admixture combinations that were found to be useful for antifreeze concrete.
(The FHWA project report [ERDC/CRREL TR-04-2] lists eight admixture
combinations that provided antifreeze protection; however, the WRG IV and MB III
combinations were not included in this ETL because the former mix caused concrete
to gain strength too slowly and the latter tended to disentrain air from the concrete.)
The six admixture combinations were developed after numerous trials to yield
concrete with the desired –5°C freezing point, reasonable transit life, and good
jobsite workability. Of all the commercial admixtures marketed today, these six
combinations are not the only possible antifreeze combinations, but they were
chosen because they performed well. They are not even the only possibilities when
using the products from the two companies chosen for this study. However, Table 1
should serve as the basis for proportioning antifreeze concrete mixtures until more
information becomes available. The admixtures in Table 1 were tested in laboratory
concretes containing 392 kg/m3 cement and the w/c ratios shown.
5
Table 1. Recommended Dosages of Six Combinations of Commercial Admixtures
to Protect Fresh Concrete Against Freezing.
Product Name
Admixture Dosage
W.R. Grace & Co.
I†
II
III
Mira 70 (mL/100 kg cement)
780
585
390
ADVA Flow (mL/100 kg)
195
98
65
DCI (L/m3 concrete)
30

30
DCI-S (L/m3)

30

6.52
6.52
6.52
Eclipse Plus
(percent of cement weight)


1
Darex II AEA (mL/100 kg)
60‡
30‡
20
0.442
0.452
0.442
PolarSet (L/100 kg)
w/c ratio
Master Builders, Inc.
I
II
IV
Polyheed 997 (mL/100 kg)
780
780
780
Glenium 3000 NS (mL/100 kg)
195
195

Rheocrete CNI (L/m3)
30
30
30
Pozzutec 20 (L/100 kg)
5.87
5.87

Pozzutec 20+ (L/100 kg)


5.87
Pozzolith100-XR (mL/100 kg)

65

MB-VR Standard (mL/100 kg)
40‡
20‡
20
0.430
0.431
0.390
w/c ratio
* The performances received with these admixture dosages may vary
from cement to cement.
† Roman numerals correspond to six admixture combinations
developed for this ETL.
‡ It may be necessary to redose with an air-entraining admixture
(AEA) at the jobsite
6
10. Proportioning Antifreeze Concrete. Until one becomes familiar with the process
for formulating CWAS, the recommended approach to proportioning antifreeze concrete
mixtures is to begin with a standard concrete used in warm-weather construction and
that is workable, durable, and strong. The procedure to convert a standard concrete into
antifreeze concrete is to choose one of the admixture combinations from Table 1, verify
that it can produce a –5°C freezing point, and conduct field trials to determine the
optimum admixture dosing sequence to achieve target slump, air content, and working
time.
10.1. Step 1: Select Standard Concrete. As described in paragraph 9.2.4, select a
concrete mixture that contains at least 363 kg/m3 cement, has a 0.45 or lower w/c
ratio, and is workable. Five standard concretes were selected for field demonstration
in the FHWA project (ERDC/CRREL TR-04-2) as shown in Table 2. Each concrete
in Table 2 typically had a 100-millimeter slump immediately after mixing. This ETL
will follow the conversion and subsequent demonstration of the Littleton, New
Hampshire mix.
Table 2. Mixture Proportions of Five Warm-Weather Concretes Selected for
Conversion to Antifreeze Concretes, and Subsequent Field Evaluation.
Littleton,
NH
Rhinelander,
WI
North
Woodstock,
NH
West
Lebanon,
NH
Concord,
NH
Type I
−
476
−
−
−
Type I/II
−
−
−
392
−
Type II
392
−
392
−
392
Coarse—19mm
1042
950
1050
1083
1050
Coarse—38mm
−
179
−
−
−
680
780
699
804
699
Darex II
78
−
233
97
233
Catexol AE260
−
252
−
−
−
Ingredient
Cement (kg/m3)
Aggregate
(kg/m3)
Fine—sand
AEA* (mL/m3)
7
Littleton,
NH
Rhinelander,
WI
North
Woodstock,
NH
West
Lebanon,
NH
Concord,
NH
0.514
−
−
0.512
−
Catexol 2000N
−
1.868
−
−
−
WRDA-Hycol
−
−
0.766
−
0.766
0.436
0.413
0.444
0.436
0.444
Ingredient
Water-reducing
admixture
(L/m3)
Mira 70
w/c
* Air-entraining admixture
10.2. Step 2: Verify Freezing Point.
10.2.1. To verify that a –5°C freezing point is achievable it is necessary to
calculate the percent solids from the admixtures by weight of free water in the
concrete. The freezing point can then be determined from Figure 1, which shows
the relationship between solids content and initial freezing point. These data can
be used to predict freezing points for antifreeze concretes made with the
admixture combinations in Table 2. Similar curves must be developed for other
admixture combinations. Each graph represents one admixture combination from
Table 1: a⎯WRG I; b⎯WRG II; c⎯WRG III; d⎯MB I; e⎯MB II; f⎯MB IV.
8
0
-1
-2
a.
Freezing Point - C
Freezing Point - C
0
-1
y = –23.789x – 1.4609
-3
-4
-5
-6
-7
b.
-2
y = –23.452x – 1.4954
-3
-4
-5
-6
-7
-8
10%
12%
14%
16%
18%
20%
22%
-8
10%
24%
12%
14%
0
0
-1
-1
-2
c.
y = –24.296x – 1.1436
-3
-4
-5
-6
-7
-8
10%
12%
14%
16%
18%
20%
22%
22%
24%
y = –32.969x + 0.8118
-3
-4
-5
-6
-8
10%
24%
12%
14%
16%
18%
20%
22%
24%
22%
24%
Solids
0
e.
-1
y = –33.78x + 0.9778
-2
Freezing Point - C
Freezing Point - C
20%
-7
0
-3
-4
-5
-6
-7
-8
10%
18%
d.
-2
Solids
-1
16%
Solids
Freezing Point - C
Freezing Point - C
Solids
f.
y = –29.749x – 0.3347
-2
-3
-4
-5
-6
-7
12%
14%
16%
18%
20%
22%
-8
10%
24%
Solids
12%
14%
16%
18%
20%
Solids
y = freezing point
x = solids content, decimal form
Figure 1. Relationship Between Weight Of Solids Contributed by Admixtures In
Percent by Weight of Free Water and Initial Freezing Point of Freshly Mixed
Concrete.
9
10.2.2. Equation 1 calculates the weight of solids contributed to the concrete
when the addition rate of admixtures is based on cement content1. Table 3
provides needed information about each admixture.
Equation 1:
Weight of solids = Dosage × Cement Factor × Sp. gr. × Solids Content
Table 3. Physical Values of Individual Admixtures
Admixture
Solids Content (%)
Specific Gravity
Mira 70
23.5
1.1
Adva Flow
31.2
1.057
33
1.273
DCI-S
32.5
1.275
PolarSet
41.1
1.35
Eclipse Plus
100
0.945
Darex II
12
1.005
47.5
1.270
30
1.08
Rheocrete CNI
32.5
1.295
Pozzutec 20
49.5
1.38
Pozzutec 20+
40
1.35
Pozzolith 100-XR
47
1.22
MB-VR
13
1.038
DCI
Polyheed 997
Glenium 3000 NS
10.2.3. The Littleton, concrete used the WRG II admixture combination. Mira 70
was dosed at 585 mL/100kg of cement (Table 1). Substituting this value, a
specific gravity of 1.1, a solids content of 23.5% (Table 3), and a 392 kg/m3
cement factor (Table 2) into equation 1 yields 0.593 kilogram solids (1.93
kilograms water), as shown below. This is the weight of solids that Mira 70
contributes to a cubic meter of concrete. In like manner, Adva Flow, PolarSet,
and Darex II contribute 0.127 kilogram, 14.181 kilograms, and 0.035 kilogram
solids,2 respectively (0.280 kilogram, 20.321 kilograms, and 0.257 kilogram
water, respectively).
585mL
× 392 kg cement ×1.1× 0.235 = 0.593kg/m 3
100 kg cement
1
The weight of water contributed by each admixture can be calculated by simply substituting [1-solids content] for the solids
content.
2
An extra 44 mL/100 kg of Darex II was added at the jobsite.
10
10.2.4. For admixtures whose addition rates are based on concrete volume, the
weight of solids is calculated by Equation 2. According to Table 1, DCI-S was
dosed at a rate of 30 L/m3. Substituting this value, a specific gravity of 1.275 and
a solids content of 32.5% into Equation 2 yields 12.431 kilograms solids (25.819
kilograms water) per cubic meter of concrete.
Equation 2:
Weight of solids = Dosage × Sp. gr. × Solids Content
30 L
×1.275 × 0.325 = 12.431 kg/m3
3
m
10.2.5. The percent solids contributed by WRG II to the Littleton concrete is
equal to the weight of solids contributed by all the admixtures divided by the
theoretical free water content3 of the concrete (170.9 kilograms = w/c ratio ×
cement factor [Table 2]). In this case the WRG II combination was designed to
contribute 16 percent solids by weight of water (27.367 kilograms solids/170.9
kilograms water). (The amount of water the ready-mix plant would add into the
mixture would be 122.3 kilograms [170.9 kilograms – 48.6 kilograms]).
10.2.6. Finally, the freezing point of this concrete is found to be –5.2°C (Fig. 1.b).
In a perfect world, each admixture could be reduced to raise the freezing point to
–5°C. However, because it is difficult to accurately control moisture when
batching ready-mix concrete, it is best to slightly overshoot the target freezing
point.
10.3. Step 3: Trial Batches.
10.3.1. Three approaches for dosing the admixtures into the concrete were
devised to ensure that an acceptable concrete could be produced. Because
variations in jobsite conditions and concrete-making materials affect concrete
performance, it is strongly recommended that trial tests be done to determine
which approach works best before placing concrete on a job. Each approach is
detailed in Attachment 1 and briefly described below. Control of workability and
air content are concerns to address. As discussed in paragraph 13.2, the ability
to measure the freezing point of the concrete needs to be refined.
10.3.2. The first approach is to mix all ingredients at the ready-mix plant. The
advantage of this approach is that once the truck leaves the plant no further effort
is needed to prepare the concrete. However, because the tested antifreeze
concretes tended to lose slump rapidly, future antifreeze concretes should have a
high initial slump to accommodate any loss during transit and still be workable
during placement. In this ETL, initial slump values of 150 to 225 millimeters
3
Includes water from aggregate, water from admixtures, and water added into the mixture.
11
yielded acceptable results during placement. Slowly agitating the concrete while
in transit was helpful for maintaining slump. The initial air content should also be
high to account for air lost during transit, as some admixtures tend to be harsh on
entrained air. It was found that initial air contents between 8 percent and 10
percent produced target value air contents during placement. Re-dosing the mix
at the jobsite with plasticizer or air-entraining admixture or both may be
necessary with some mixes. This first approach is best when the haul time is no
more than 45 minutes and the concrete can be immediately dispensed within 20
minutes.
10.3.3. The second approach is to add half of the admixtures into the concrete at
the ready-mix plant and the rest at the jobsite. The non-accelerating admixtures,
if any, should be added first. This approach requires that admixtures be
transported to the jobsite and pumped into the truck. Though requiring extra
effort, it provides more assurance that the concrete will be workable, as slump
loss is less of a factor until the final admixture is added into the concrete. Redosing the mix at the jobsite with an air-entraining agent may not be necessary
with this approach. However, working time tends to decrease very rapidly once
the last admixture is added, so the concrete must be dispensed quickly—usually
within 30 minutes. This approach may allow haul times of as long as one hour.
10.3.4. The third approach is to mix all of the admixtures into the concrete at the
jobsite. This approach requires the most effort, but the admixtures do not have to
be dosed until the construction crew is ready for the concrete and it allows 45
minutes or longer for working time. Because the concrete must leave the readymix plant with a very low w/c ratio to account for the water in the admixtures, the
air-entrained admixture (AEA) is best added into the mix at the jobsite.
11. Field Demonstrations.
11.1. Once trial batches prove that an acceptable antifreeze concrete can be
produced, it is ready for full-scale use. The Littleton demonstration project
(ERDC/CRREL TR-04-2) is used as an example.
11.2. Because the estimated haul time to the Littleton project was 30 minutes, it was
felt that Approach 2, paragraph 10.3.3, could be followed. Consequently, all but the
PolarSet and some of the AEA were withheld from the batch until the truck arrived at
the jobsite (Table 4). Once there, a quick visual check revealed that the concrete
had a slump of about 100 millimeters, which was higher than the expected 70
millimeters determined by the trial tests. This immediately suggested that the water
content of the mixture was high. However, because the solids content was designed
to be higher than necessary, no adjustments were made and the PolarSet and AEA
were added into the truck. The resulting slump, measured with a slump cone, was
203 millimeters. Because the mix was cohesive, the concrete could have been
immediately placed into its forms, but it was elected to slowly agitate it for 20
minutes at 2 to 4 revolutions per minute (rpm) until it began to lose slump. Once that
12
happened, the concrete was placed during the next 40 minutes, dropping to the
target slump of 100 millimeters. The freezing point of the concrete was unknown at
this point4.
Table 4. Dosing Schemes Used in the Five Field Studies.
Littleton,
NH
Rhinelander,
WI
North
Woodstock,
NH
West
Lebanon,
NH
Concord,
NH
29.84
[18.5]
−
−
−
−
[11.3]
−
−
−
[25.56]*
[30.9]
−
−
−
Rheocrete CNI
−
−
[29.8]
29.8
[29.8]
Pozzutec 20+
−
−
[23.0]
23.0†
[23.0]
116 [170]
−
−
−
−
Catexol AE260
−
407 [156]
−
−
−
MB-VR
−
−
[38.9]
78
[38.9]
Mira 70
2.293
[3.74]
−
−
−
Adva Flow
0.389
[0.934]
−
−
−
−
−
[2.289]
2.289
[2.304]
Ingredient
Freezing point
and accelerating
admixture (L/m3)
DCIS
DCI
PolarSet
AEA‡ (mL/m3)
Darex II
Water-reducing
admixture (L/m3)
Polyheed 997
* Brackets denote jobsite additions; all other numbers indicate ingredients added into
the concrete at the ready-mix plant.
† This admixture was pumped into the truck but not mixed into the concrete until the
truck reached the jobsite.
4
This pooled-fund project did not include as one of its goals the measurement of freezing point. However, there is no way to know
that the concrete fully met expectations unless such a measurement is made. An expedient measurement method is described in
paragraph 13.2 and in more detail in ERDC/CRREL TR-04-2. Our field measurements showed that the Littleton concrete’s freezing
point was –5.2°C. It was fully acceptable. Attachment 2 shows that freezing point measurements can used to backcalculate actual
w/c ratios.
13
‡ Air-entraining admixture
11.3. The other field tests, except for the West Lebanon test (ERDC/CRREL TR-042), dosed most admixtures into the concrete at the jobsite. With today’s admixture
pumps this was not viewed as a difficult task. This scheme provided full assurance
that the load would not stiffen during transit and that the concrete could be easily
adjusted on site to give the desired workability. The West Lebanon test tried an
interesting variation: all admixtures were dosed at the ready-mix plant, except that
one admixture was not mixed into the concrete—it was merely pumped into the truck
and then mixed once the truck arrived at the jobsite. This scheme avoided the need
for admixture pumps at the jobsite altogether and provided highly workable concrete.
12. Production and Placement. Antifreeze concrete mixtures are sensitive to water
content when it comes to freeze protection. They tend to lose slump very rapidly, are
sticky to finish, and require close control of when admixtures are added into the mixing
process. However, they show no tendency to bleed when properly batched, so the
finishing operation can be started almost immediately after the concrete is consolidated
and leveled. Even at very low slumps, the concretes have responded well to vibration.
12.1. Ready-Mix Plant.
12.1.1. The ready-mix plant should be within a 30-minute drive of the jobsite to
allow time for positioning the truck and making any final adjustments to the
concrete (longer drives are permissible if all admixtures are dispensed at the
jobsite). Because of the need to create a concrete of a certain freezing point,
accurate moisture control is critical. Ready-mix plants should have moisture
meters to continuously monitor changes in the sand as it is batched. The
moisture content of the coarse aggregate should be measured at the start of
each day and whenever the slumps start to deviate. For example, in the FHWA
study (ERDC/CRREL TR-04-2), moisture contents were obtained from grab
samples taken from the sand and coarse aggregate piles a few hours before
batching. However, it was found that accurately measuring aggregate moisture
proved difficult. In the first three field trials, the actual water content of the
concretes was higher than designed. Snow in the aggregate piles was
considered responsible for errors in measurement.
12.1.2. In response to these difficulties, the aggregate moisture contents in the
final two field trials were assumed to be somewhat greater than the grab sample
measurements indicated. This led to both concretes containing lower w/c. In
these instances, extra water could have been added into the mix.
12.1.3. Unless better moisture control is possible, it is recommended that the
ready-mix plant withhold up to 25 L of batch water from every cubic meter of
concrete. Once all admixtures are mixed into the concrete, then the withheld
water should be carefully mixed into the concrete until it attains the desired slump
as developed in the trial batches.
14
12.2. Truck Driver.
12.2.1. Before batching, truck drums should be reversed for numerous rotations
to ensure no wash water is in them. It is also recommended not to wash down
the fins after loading to avoid getting extra water into the drum. If this is not
possible, then the ready-mix plant should withhold extra water to account for the
wash water. The truck driver should demonstrate his washing technique in
several trials so the amount of wash water can be reasonably estimated. Until
better data becomes available, it can be estimated that 10 to 15 L of water are
used to wash down fins.
12.2.2. The same driver should be used during both the trial batches and the
actual jobsite placement. If the truck has a resistance meter, the driver should
observe its readings during several parts of the trials to know when the concrete
has good slump and, more importantly, when it has begun to stiffen. Adding
water at this point will loosen the mixture, but waiting too long may create
problems.
12.2.3. During transit, the mix should be slowly agitated at 2 to 4 rpm if trial
batches indicate rapid stiffening of the mix. Once at the jobsite and with the truck
positioned it is recommended that the concrete be mixed 15 revolutions if no
admixtures are added into the truck and 30 revolutions if admixtures are added.
12.2.4. Though not experienced in the FHWA study (ERDC/CRREL TR-04-2),
during field placements the last part of the load may become difficult to
discharge; if this happens, batch size should be reduced. Typically, it is
recommended that the maximum batch size be limited to half—but no more than
two-thirds—of the mixing capacity of the truck. Half-capacity is recommended for
the first few loads until experience is gained with the particular concrete being
used. Batch size may be increased if conditions warrant.
12.2.5. The concrete is made with cold water, but it is important to have hot water
in the truck’s water tank. The hot water will prevent the truck’s hoses and nozzles
from freezing up during cleanup operations.
12.3. Job Supervisor.
12.3.1. Successful use of antifreeze concrete requires close coordination
between the ready-mix plant, the concrete truck driver, and the concrete
construction crew. A pre-construction meeting should be held with these key
individuals to inform them how the concrete should behave during batching,
transport, and placement—and particularly how quickly the concrete can stiffen.
12.3.2. Because antifreeze concrete stiffens very rapidly, concrete that is not
used should be temporarily disposed of at a predetermined jobsite location. If this
15
is not possible, instruct the truck driver to water down the waste concrete until it
has a soupy consistency; otherwise the paste may harden onto the truck’s fins,
requiring extra effort to clean out the drum. Alternatively, one could use a
hydration-controlling admixture to prevent the concrete from hardening on the trip
back to the ready-mix plant.
12.3.3. Trial mixing is recommended because conditions and materials vary with
location. Several 2- to 4-cubic-meter trial batches should be made to confirm
mixture proportions and to give participants a feel for how the concrete will
behave during each stage of the process. The trial batches should be used to
define the proper amounts, sequencing and timing of the admixtures based on
their response to the particular cement being used. Obtain periodic slump
measurements, air content, and concrete temperature readings to verify that the
concrete mixture will behave as desired throughout the time of placement. The
trial batches provide the opportunity to experiment with different admixture
dosing schemes for training the batch plant, truck driver, and job supervisor in
working with antifreeze concrete. As with all new construction methods, there will
be a learning curve.
12.3.4. Though the antifreeze concrete produced as recommended in this ETL is
capable of resisting freezing to –5°C, it is not possible at this point to forecast to
what outdoor temperature this translates. The thermal profile of curing concrete
is a function of the mixture proportions, its geometry, the amount of wind and
solar radiation, and the type of formwork. Until the hydration process is better
characterized for low-temperature application, the best guidance that can be
provided is that antifreeze concrete can be placed when the air temperature is at
least –20°C and rising. The concrete was placed in sections as thin as 140
millimeters on gravel when the air temperature started out at approximately –
20°C, rose to about –10°C over the next six hours, and then dropped to below –
20°C that night. The concrete section was covered with a 25-millimeter-thick
insulation blanket.
12.3.5. When removing formwork, you should consider ensuring that the
concrete is not exposed to a large thermal gradient. Follow guidance provided in
ACI 306-R88, Cold Weather Concreting.
12.4. Placing and Finishing.
12.4.1. The substrate against which fresh concrete is to be placed must be free
of ice and snow. It is permissible for its temperature to be below freezing, but all
sources of excess water must be removed.
12.4.2. The behavior of antifreeze concrete in its plastic state is very similar to
that of high-cement-content, low-w/c-ratio concrete. Like conventional concrete, it
is sticky to finish but tends to lose its stickiness as slump decreases. The
objective is to produce a uniformly dense cross-section that will be suitable for
16
the service conditions. The serviceability of a concrete surface is often affected
by the finishing procedures. It is important for durability and wear resistance not
to overwork the surface.
12.4.3. The concrete should be continuously placed within 20 to 30 minutes or
according to the working time determined during the trial batching tests. As the
concrete is placed, immediately strike off excess concrete to bring the surface to
proper elevation; this can be done either mechanically or by hand with a
straightedge. Following screeding, bull float the surface to embed aggregate and
smooth the surface. Be careful not to seal the surface. Some tearing when using
a magnesium float is not undesirable. Wait until the concrete can be walked upon
without leaving footprints more than 0.25-inch deep before floating. Floating too
early is likely to cause dusting and crazing because floating tends to bring
cement paste to the surface. A float finish followed by scratching the surface with
a broom is often recommended for flatwork exposed to outdoor conditions.
Stopping at this point provides a durable, uniform, non-skid surface and avoids
problems caused by over-troweling.
12.4.4. If troweling is desired, a single troweling pass, rather than several, is
preferred. The purpose of troweling is to produce a hard, smooth surface.
Minimal troweling should leave a somewhat rough surface without creating a
weakness at that level. Overworking the surface can lead to a weak surface layer
caused by working in too much water and a surface susceptible to frost action
because of a diminished air content caused by too much mechanical action.
12.4.5. Once finishing is completed, exposed surfaces must be protected against
drying out or otherwise strength and wear resistance will suffer. The surface
should be covered with a sheet of plastic or a spray-on curing compound (or
both) as soon as the surface is tack-free. This happens very quickly. Wrinkle-free
application of the plastic sheet helps to ensure a uniformly colored surface.
Cover protruding metal with insulation since metal can act as a heat sink to
freeze surrounding concrete. Be sure to insulate the ends of metal form ties.
12.4.6. If sawed joints are necessary, sawing should be done when test cuts
have sharp, clean edges.
13. Quality Verification.
13.1. Strength Gain Estimates.
13.1.1. A critical question to be answered for any concrete structure is “When will
it be ready for use?” There are a number of tests, both destructive and nondestructive, that can be used to determine this, but the maturity method is
favored. A detailed discussion of how to develop maturity curves is provided in
the FHWA project report (ERDC/CRREL TR-04-2). Common practice is to
develop a maturity curve from test concrete having the same mixture proportions
17
as the intended job concrete before starting the construction project. In practice
this usually does not happen because it is difficult to plan ahead for most
projects, particularly if they are small. For many projects it is important to put the
structure into service as soon as possible. Thus, it is recommended that a
maturity curve be developed from the job concrete as it is being placed.
13.1.2. Before placement, thermocouples should be placed at several locations
where it is important to know what the strength is at any given time. Then, during
placement, numerous cylinders should be cast from the fresh concrete. Typically,
about one-third of the load should be placed before cylinders are cast. Install
thermocouples into several cylinders. Then, allow the cylinders to cure at any
convenient temperature, provided their internal temperature does not drop below
–5°C, and test them, three at a time, for compressive strength. Store the
cylinders indoors. The time-temperature factor recorded from the cylinders is
then plotted against cylinder strength. This plot is then used to estimate in-situ
strength based on the time-temperature factor recorded from the structure.
Temperature data collection continues until it is determined that the structure is
safe to put into service (see the FHWA project report [ERDC/CRREL TR-04-2]
for more detail).
13.2. Freezing Point Measurement.
13.2.1. It is critical to know the temperature at which water in fresh concrete
freezes. This temperature is affected by the type and concentration of solids
dissolved in the water. This is an important parameter to know when concrete
arrives at a construction site; admixtures are dosed knowing the moisture content
of aggregate. However, the actual w/c ratio in ready-mix concrete can vary by
several percent due to changes in aggregate moisture, measurement
inaccuracies, and the tendency of contractors to add water at the site. Thus, the
final freezing point of the concrete could be considerably different from desired.
Unfortunately, there are no commercial devices to determine the freezing points
of concrete in the field.
13.2.2. Until a better method is developed, CRREL has developed a rudimentary
way to measure the freezing point of fresh concrete in the field. Because this
measurement takes about 30 minutes to complete, it is most useful when used in
conjunction with the trial batch operation. It is envisioned that a relationship
between freezing point and slump would be the most useful product created for
the concrete mixture selected for the job.
13.2.3. The freezing point method is described in the FWHA project report
(ERDC/CRREL TR-04-2). However, in brief, it consists of making several 50.8by 101.6-millimeter cylindrical samples, installing thermocouples, and placing
them in a chest containing dry ice. Experience shows that these cylinders might
drop below their freezing point within 20 minutes of being placed in the chest.
Having the data plotted real-time onto a laptop computer facilitates rapid
18
determination of the freezing point. Freezing points should be determined at
various slumps (w/c ratios). The freezing point is identified as the location on the
temperature vs. time plot (Figure 2) from the cylinders being cooled where the
initial slope of the cooling curve (the mostly linear portion above 0°C) suddenly
changes. The resulting relationship between freezing point and slump can then
be used to estimate the as-dispensed freezing point of the concrete at the
jobsite, provided the admixtures were used in the same doses as found in the
trial batches.
Temperature - C
30
20
FREEZING POINTS
10
Tap Water
0
Control Concrete
-10
Antifreeze Concrete
-20
0
100
200
TIME - Minutes
300
d
/P
lfF
d/ h
400
II/F
i
i t /F
i
tf
h
t /
i 11
Figure 2. Cooling Curves for Water, Control Concrete, and Antifreeze Concrete.
14. Contact. Recommendations for improvements to this ETL are encouraged and
should be furnished to HQ AFCESA/CESC, 139 Barnes Drive, Suite 1, Tyndall AFB, FL
32408-5319, Pavements Engineer, DSN 523-6334, commercial (850) 283-6334,
internet AFCESAReachBackCenter@tyndall.af.mil.
LESLIE C. MARTIN, Colonel, USAF
Director of Engineering Support
4 Atch
1. Trial Batching
2. Determining the W/C Ratio From
Freezing Point Measurements
3. Conversion Factors
4. Distribution List
19
TRIAL BATCHING
A1.1. The trial batches will determine the best timing for adding admixtures into the
concrete. To reasonably duplicate the performance of concretes batched for jobsite
application, use a minimum batch size of 2 to 4 cubic meters. Use cold water for all
batching operations to create an initial concrete temperature of 10° ± 5°C. Higher
temperatures tend to cause antifreeze mixtures to stiffen too rapidly and lower
temperatures are difficult to achieve. The goal is to produce a concrete with acceptable
slump and air content at the jobsite. Jobs with short haul times and cement content less
than 400 kg/m3 can have all admixtures added into the concrete at the ready-mix plant,
whereas longer haul times or higher cement content might require that all the
admixtures be added at the jobsite. Moisture control is critical to obtain a desired
freezing point—determine moisture content of sand and coarse aggregate before the
start of each day and as needed during the day. Reverse the drum on the truck to
empty wash water from previous cleaning operations before making a trial batch.
A1.2. The three approaches developed for dosing admixtures into concrete that
performed well for this ETL are presented in this attachment. The times shown simulate
the time spent at the ready-mix plant, in transit, and on the jobsite, and are based on
performance requirements established at the beginning of the study: 45-minute transit
time and 20- to 30-minute jobsite working time. All mixing and testing of trial batches
can take place at the ready-mix plant.
Atch 1
(1 of 5)
Approach 1. Dose all admixtures at the ready-mix plant.
Time (hr:min)
Location
Operation
Jobsite
Transit
Ready-mix plant
Add cold mixing water into drum
Add sand, coarse aggregate and cement into
drum, with AEA* on sand
Mix for 2 minutes
With mixer slowly agitating (2–4 rpm), add one
of the following admixtures into the drum: DCI,
DCI-S, or Rheocrete CNI†
Mix for 1 minute
Back load up to top of drum and add either
Mira 70 or Polyheed 997†
Mix for 1 minute
Back load up to top of drum and add either
Adva Flow or Glenium 3000 NS†
Mix 1 minute
With mixer slowly agitating (2–4 rpm), add one
of the following admixtures into the drum:
PolarSet, Pozzutec 20, or Pozzutec 20+†
Mix for 3 minutes, then agitate at 2–4 rpm
thereafter
Obtain slump and air content
Obtain slump and air content
Obtain slump and air content
Re-dose with plasticizer and AEA if needed
(back up load)
Obtain slump, air content, freezing point and
cylinders
Immediately add water into drum and discard
remaining concrete
* Air-entraining admixture.
† Depends on the selected admixture combination.
‡ Timing starts when water hits cement.
Start − End
Total
0:00
0:00‡
0:00 – 0:02
0:02 – 0:05
0:05 – 0:06
0:06 – 0:08
15
min
0:08 – 0:09
0:09 – 0:11
0:11 – 0:12
0:12 – 0:15
0:15 – 0:18
0:20 – 0:30
0:35 – 0:45
0:50 – 1:00
45
min
1:00 – 1:20
1:40 – 1:50
30
min
Comments: Because all admixtures are mixed into the concrete at the ready-mix plant,
the concrete may lose slump and air rapidly while in transit to the jobsite. Travel times in
excess of 20 minutes may cause unacceptable slump loss with some cement, though in
this study slumps could be recovered with extra plasticizer dosed into the truck after
traveling up to 45 minutes. Working times of 20 to 30 minutes at the jobsite can be
expected with this approach.
Atch 1
(2 of 5)
Approach 2. Dose some admixtures at the ready-mix plant and the
rest at the jobsite.
Time (hr:min)
Location
Operation
Jobsite
Transit
Ready-mix plant
Add cold mixing water into drum
Add sand, coarse aggregate and cement into
drum, with AEA on sand
Mix for 2 minutes
With mixer slowly agitating (2–4 rpm), add
one of the following admixtures into the
drum: DCI, DCI-S, or Rheocrete CNI*
Mix for 1 minute
Back load up to top of drum and add either
Mira 70 or Polyheed 997*
Mix for 1 minute
Back load up to top of drum and add either
Adva Flow or Glenium 3000 NS*
Mix for 3 minutes, then agitate at 2–4 rpm
thereafter
Obtain slump and air content
Obtain slump and air content
Obtain slump and air content
With mixer slowly agitating (2–4 rpm), add
one of the following admixtures into the
drum: PolarSet, Pozzutec 20, or Pozzutec
20+*
Mix for 3 minutes
Obtain slump and air content
Re-dose with plasticizer and AEA if needed
(back up load)
Mix for 2 minutes
Obtain slump, air content, freezing point and
cylinders
Immediately add water into drum and discard
remaining concrete
* Depends on the selected admixture combination.
† Timing starts when water hits cement.
Start − End
Total
0:00
0:00†
0:00 – 0:02
0:02 – 0:05
0:05 – 0:06
11
min
0:06 – 0:08
0:08 – 0:09
0:09 – 0:11
0:11 – 0:14
0:16 – 0:26
0:31 – 0:41
0:46 – 0:56
45
min
1:00 – 1:03
1:03 – 1:06
1:07 – 1:17
1:18 – 1:21
30+
min
1:21 – 1:23
1:25 – 1:35
Comments: Because some admixtures are withheld from the concrete at the ready-mix
plant, the concrete may not lose air as rapidly during transit as in Approach 1. However,
the mix may need higher doses of plasticizer to keep it mobile during transit. Transit
Atch 1
(3 of 5)
times of 45 minutes or more should be possible with this approach, particularly if no
accelerating admixtures are in the mix. Working times might still be limited to 30
minutes at the jobsite, as slump loss is very rapid once the final admixtures are mixed
into the concrete.
Approach 3. Dose all admixtures at the jobsite.
Time (hr:min)
Location
Operation
Start − End
Readymix plant
Add cold mixing water into drum
0:00
Add sand, coarse aggregate and cement into
0:00†
drum
Mix for 2 minutes
0:00 – 0:02
Transit
Agitate at 2–4 rpm
0:02 – 0:47
Jobsite
With mixer slowly agitating (2–4 rpm), add
one of the following admixtures into the drum:
DCI, DCI-S, or Rheocrete CNI*
Mix 1 minute
Back load up to top of drum and add either
Mira 70 or Polyheed 997*
Mix 1 minute
Back load up to top of drum and add AEA that
has been sprinkled into a bucket containing
approximately 5 kg sand
Mix 2 minutes
With mixer slowly agitating (2–4 rpm), add
one of the following admixtures into the drum:
PolarSet, Pozzutec 20, or Pozzutec 20+*
Mix 3 minutes
Obtain slump and air content
Back load up to top of drum and add either
Adva Flow or Glenium 3000 NS*, if needed;
mix 1 minute.
Obtain slump, air content, freezing point and
cylinders
Add water into drum and discard remaining
concrete
* Depends on the selected admixture combination.
† Timing starts when water hits cement.
Atch 1
(4 of 5)
Total
2 min
45
min
0:50 – 0:53
0:53 – 0:54
0:54 – 0:56
0:56 – 0:57
0:57 – 0:59
0:59 – 1:01
1:01 – 1:04
1:04 – 1:07
1:08 – 1:18
1:18 – 1:21
1:23 – 1:30+
40+
min
Comments: Because all admixtures are added at the jobsite, there is little concern
about stiffening problems during delivery. The admixtures do not need to be added into
the mix until the construction crew is ready. However, because the concrete is initially
prepared with a very low w/c ratio, the AEA should be added at the jobsite. Transit times
can easily exceed 45 minutes. There may be no need for a super-plasticizer (Adva Flow
or Glenium). Working times at the jobsite can last 45 minutes or longer.
Atch 1
(5 of 5)
DETERMINING THE W/C RATIO FROM FREEZING POINT MEASUREMENTS
A2.1. Because the amount of admixtures and cement in most ready-mixed concretes
are more accurately known (compared to water contents), a freezing point
measurement can be used to estimate the actual w/c ratio of the concrete. For example,
if the MB II admixture combination were used in a concrete that was found to have a
–4.5°C freezing point, it is immediately known from Figure 1.e. that the water in that
concrete contains 16.2 percent solids. This value, along with the amount of cement and
the dosage of admixtures used to make the concrete, are all that is needed to determine
the w/c. The dosage values are converted to weights of solids by following the
procedure outlined in paragraph 10.2. Assuming that the admixtures contributed 35.38
kg/m3 solids, there would be exactly 218.39 kg/m3 water (35.38/16.2 percent) in the
freshly mixed concrete. If the batch ticket indicated that 496 kg/m3 cement were used in
the concrete, a 0.44 w/c ratio is calculated. At this point the concrete could be rejected,
if its freezing point was not met, or more admixture could be added into the concrete, or
the structure could be thermally protected to prevent freezing.
Atch 2
(1 of 1)
CONVERSION FACTORS
TO CONVERT
TO
DIVIDE BY
LENGTH
millimeters (mm)
centimeters (cm)
meters (m)
meters (m)
meters (m)
kilometers (km)
inches (in)
inches (in)
inches (in)
feet (ft)
yards (yd)
miles (mi)
25.4
2.54
0.0254
0.3048
0.9144
1.60948
AREA
square millimeters (mm2)
square centimeters (cm2)
square meters (m2)
square meters (m2)
square meters (m2)
square kilometers (km2)
square kilometers (km2)
square inches (in2)
square inches (in2)
square inches (in2)
square feet (ft2)
square yards (yd2)
square miles (mi2)
acres
645.16
6.4516
0.00064516
0.09290
0.83613
2.59043
0.00404
VOLUME
cubic millimeters (mm3)
cubic centimeters (cm3)
cubic meters (m3)
cubic meters (m3)
cubic inches (in3)
cubic inches (in3)
cubic feet (ft3)
cubic yards (yd3)
16,387
16,487,000
0.028317
0.764559
MASS
kilograms (kg)
pounds (lb)
0.45359
FORCE
Newtons (N)
pounds (lbf)
4.44822
STRESS
kiloPascals (kPa)
pounds per square
6.89476
Atch 3
(1 of 1)
DISTRIBUTION LIST
DEPARTMENT OF DEFENSE
Defense Commissary Agency
(1)
Design and Construction Division
2250 Foulois St., Suite 2
Lackland AFB, TX 78236
AAFES
ATTN: RE-C
PO Box 660202
Dallas, TX 75266-0202
(1)
Construction Criteria Base
National Institute of Bldg Sciences
Washington, DC 20005
(1)
SPECIAL INTEREST ORGANIZATIONS
Information Handling Services
15 Inverness Way East
Englewood, CO 80150
(1)
Atch 4
(1 of 1)